PROJECT SUMMARY/ABSTRACT Developmental language disorder (DLD) affects 6-8% of children and has impacts on academic, social, and economic outcomes throughout life. Early identification of children at risk for DLD is critical to improving long- term outcomes. However, for the large number of bilingual children in the United States, early signs of DLD can be difficult to recognize. A key challenge in identifying bilingual toddlers at risk for DLD is that current measures cannot differentiate between delays in accumulated language knowledge ? which can result from reduced exposure to each language ? and ?true? impairments in language learning and processing that are likely to result in persistent problems. Measures of language processing offer a promising approach to identifying DLD in bilingual populations, as they try to capture underlying deficits while minimizing the effects of prior experience. This study will use fine-grained measures of spoken language processing to compare the performance of Spanish-English bilingual toddlers with low expressive vocabularies (?late talkers?, n=40) and typical expressive vocabularies (?typically developing?, n=40), and to examine associations between language processing and vocabulary growth over the third year of life in both groups. Children will be followed longitudinally from 24 to 36 months with measures of vocabulary (via parent report) and language processing (using eye-tracking) in Spanish and English. The overarching goal is to determine whether language processing efficiency is an early marker of DLD in bilingual toddlers. Aim 1 will use eye-tracking measures to compare the performance of typically developing and late-talking bilingual toddlers in familiar word recognition at 24 and 30 months. Aim 2 will use eye-tracking methods to compare the performance of typically developing and late-talking bilingual toddlers in novel word learning at 24 and 30 months. Aim 3 will determine the extent to which measures of language processing and vocabulary at age 2 predict trajectories of vocabulary growth across the 3rd year in both typically developing and late talking bilingual toddlers, as well as their performance in standardized tests of vocabulary knowledge at age 3. The working hypothesis is that measures of familiar word recognition and novel word learning will improve prediction of language outcomes over measures of vocabulary alone, and will help identify bilingual LTs who remain delayed at age 3. This represents the first investigation of the language processing skills of bilingual late talkers, and of the extent to which these skills predict their lexical development. Completion of study aims will inform the development of methods for early identification of DLD in bilinguals and will advance theoretical understandings of the mechanisms underlying language impairments more generally.